US20020144206A1 - Error correction apparatus for performing consecutive reading of multiple code words - Google Patents

Error correction apparatus for performing consecutive reading of multiple code words Download PDF

Info

Publication number
US20020144206A1
US20020144206A1 US10105010 US10501002A US2002144206A1 US 20020144206 A1 US20020144206 A1 US 20020144206A1 US 10105010 US10105010 US 10105010 US 10501002 A US10501002 A US 10501002A US 2002144206 A1 US2002144206 A1 US 2002144206A1
Authority
US
Grant status
Application
Patent type
Prior art keywords
code words
buffer memory
read
data
error correction
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
US10105010
Other versions
US7143331B2 (en )
Inventor
Shin-ichiro Tomisawa
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Sanyo Electric Co Ltd
Original Assignee
Sanyo Electric Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/07Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/08Error detection or correction by redundancy in data representation, e.g. by using checking codes
    • G06F11/10Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's
    • G06F11/1008Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's in individual solid state devices
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11CSTATIC STORES
    • G11C11/00Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor
    • G11C11/21Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements
    • G11C11/34Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements using semiconductor devices
    • G11C11/40Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements using semiconductor devices using transistors
    • G11C11/401Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements using semiconductor devices using transistors forming cells needing refreshing or charge regeneration, i.e. dynamic cells
    • G11C11/4063Auxiliary circuits, e.g. for addressing, decoding, driving, writing, sensing or timing
    • G11C11/407Auxiliary circuits, e.g. for addressing, decoding, driving, writing, sensing or timing for memory cells of the field-effect type
    • G11C11/408Address circuits
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11CSTATIC STORES
    • G11C7/00Arrangements for writing information into, or reading information out from, a digital store
    • G11C7/10Input/output [I/O] data interface arrangements, e.g. I/O data control circuits, I/O data buffers
    • G11C7/1006Data managing, e.g. manipulating data before writing or reading out, data bus switches or control circuits therefor
    • HELECTRICITY
    • H03BASIC ELECTRONIC CIRCUITRY
    • H03MCODING; DECODING; CODE CONVERSION IN GENERAL
    • H03M13/00Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes
    • H03M13/29Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes combining two or more codes or code structures, e.g. product codes, generalised product codes, concatenated codes, inner and outer codes
    • HELECTRICITY
    • H03BASIC ELECTRONIC CIRCUITRY
    • H03MCODING; DECODING; CODE CONVERSION IN GENERAL
    • H03M13/00Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes
    • H03M13/29Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes combining two or more codes or code structures, e.g. product codes, generalised product codes, concatenated codes, inner and outer codes
    • H03M13/2906Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes combining two or more codes or code structures, e.g. product codes, generalised product codes, concatenated codes, inner and outer codes using block codes
    • H03M13/2909Product codes
    • HELECTRICITY
    • H03BASIC ELECTRONIC CIRCUITRY
    • H03MCODING; DECODING; CODE CONVERSION IN GENERAL
    • H03M13/00Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes
    • H03M13/29Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes combining two or more codes or code structures, e.g. product codes, generalised product codes, concatenated codes, inner and outer codes
    • H03M13/2906Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes combining two or more codes or code structures, e.g. product codes, generalised product codes, concatenated codes, inner and outer codes using block codes
    • H03M13/2921Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes combining two or more codes or code structures, e.g. product codes, generalised product codes, concatenated codes, inner and outer codes using block codes wherein error correction coding involves a diagonal direction

Abstract

An error correction apparatus for performing an error correction process on digital data that is stored in a buffer memory and includes multiple code words. The device includes a memory access circuit for controlling reading and writing of the code words to the buffer memory. Operational circuits perform a syndrome calculation with each of the multiple code words read from the buffer memory. The memory access circuit consecutively reads the multiple code words from the buffer memory and distributes the code words to the operational circuits.

Description

    BACKGROUND OF THE INVENTION
  • The present invention relates to an error correction apparatus, and more particularly, to an error correction apparatus for calculating the syndrome of the data stored in a buffer memory, which includes a random access memory, to correct errors. [0001]
  • FIGS. [0002] 1(a) to 1(d) illustrate the data format of a compact disc ROM (CD-ROM) As shown in FIG. 1(a), CD-ROM data is formed from multiple sectors, each of which has 2352 bytes. Each sector includes 12 bytes of synchronizing data, 4 bytes of a header, 2048 bytes of user data, 4 bytes of error detection code (EDC) data, 8 bytes of ZERO data, and 276 bytes of error correction code (ECC) data.
  • The data excluding the 12 bytes of synchronizing data undergoes error correction with an ECC. When the error correction is performed, the 2340 bytes of data, which do not include the synchronizing data, are alternately divided into LS bytes and MS bytes. This generates data including 1170 LS bytes (FIG. 1([0003] c)) and 1170 MS bytes (FIG. 1(d)). Further, referring to FIG. 2, the 1170 LS bytes and the 1170 MS bytes are coded in two directions, a P sequence (P group) direction and a Q sequence (Q group) direction.
  • For the first 1032 bytes of data, a F parity, which consists of 2 bytes of data, is added to every 24 bytes of data in the P sequence direction. Further, 43 code words (P codes), which are coded in the P sequence direction, are configured. As apparent from FIG. 2, the position of the j[0004] th symbol for the ith code word of the P codes is represented by “i+43j (i=0, 1, . . . 42: j-0, 1, . . . 25).
  • For the first 1032 bytes of data and the 86 bytes of P parities (2 bytes ×43), a Q-parity, which consists of 2 bytes of data, is added to every 43 bytes of data. Further, 26 code words (Q words), which are coded in the Q sequence direction, are configured. The position of the j[0005] th symbol for the ith code word of the Q codes is represented by “43ii44j” mod (remainder) 1118 (i=0, 1, . . . 25: J=0, 1, . . . 42).
  • The symbols configuring the P codes and the Q codes are read from a memory to calculate a syndrome and determine whether the data includes errors. The detection of errors using the code words is performed as described below. [0006]
  • A compact disc (CD) signal processing circuit performs a decoding process, such as error correction, on the data read from a CD-ROM in the same manner as CD audio. The processed data is stored as the CD-ROM data of FIG. 1([0007] a) in a dynamic random access memory (DRAM), which is a buffer memory. FIG. 3 illustrates the CD-ROM data stored in the DRAM.
  • More specifically, with regard to the CD-ROM sector data of FIG. 1([0008] a), the 1176 bytes of data (6 bytes of data in the 12 bytes of synchronizing data and the 1170 LS bytes or MS bytes) are written along the addresses of the DRAM. FIG. 3 shows an example in which column addresses have 256 bytes. A row address is incremented in cycles of 256 bytes. The MS byLes arid LS bytes are actually written along each address in units of two bytes in the DRAM. The memory cells of the same row are connected to the same word line in the DRAM.
  • The writing of data is basically performed in accordance with the order of a data stream when the CD-ROM data is transferred (i.e., the data stream of the CD-ROM data of FIG. 1([0009] a)). However, the data is actually divided into 1170 LS bytes and 1170 MS bytes. Thus, the 6 bytes of synchronizing data and the data of FIGS. 1(c) and 1(d) are written to the memory cells starting from lower addresses. For example, referring to FIG. 3, in row 0 of sector 1, the synchronizing data is written in column 1 to column 5. Data 0 of FIGS. 1(c) and 1(d) is written to row 0 column 6. Data 1 of FIGS. 1(c) and 1(d) is written to row 0 column7.
  • The CD-ROM data of each sector is read trom the DRAM in units of code words. As an example, the reading order of the P codes will now be discussed. As shown in FIG. 4, the CD-ROM data is read in code word units of 26 bytes. The symbols forming the read code words are used to calculate the syndrome. The syndrome is set so that it has a predetermined value when each piece of data (code word) does not include an error. Thus, the calculation of the syndrome enables determination of whether a code word includes an error. [0010]
  • When it is determined that the code word includes an error, the value of the error and the location of the error are obtained based on the syndrome. Then, data is read again from the location including an error. An exclusive OR operation is performed with the read data and the value of the error to generate correct data. The erroneous data is then rewritten by the correct data. [0011]
  • After the error correction, the EDC data of FIG. 1([0012] a) is used to check whether the error correction was performed properly. The data that has undergone the error correction process and the check for erroneous corrections is transferred from the DRAM to the host computer.
  • In this manner, the code words are used to perform error correction. However, the symbols are stored in the memory cell of the DRAM at non-consecutive addresses. Thus, the time for accessing the memory cell becomes long when reading the code words. More specifically, to read a symbol from the DRAM, three operations, which include precharge, row address designation, and column address designations are necessary. Thus, three clock signals must be provided each time a symbol is read. This hinders with increasing the speed of the error correction process. Such problem occurs in an error correction apparatus that stores data temporarily in a buffer memory and then sequentially reads symbols while performing error detection or error correction. [0013]
  • SUMMARY OF THE INVENTION
  • A perspective of the present invention is an error correction apparatus for performing an error correction process on digital data that is stored in a buffer memory and includes multiple code words. The apparatus includes a memory access circuit for controlling reading and writing of the code words to the buffer memory. A plurality of operational circuits perform a syndrome calculation with each of the multiple code words read from the buffer memory, The memory access circuit consecutively reads the multiple code words from the buffer memory and distributes bile code words to the plurality of operational circuits. [0014]
  • A further perspective of the present invention is an error correction apparatus for performing an error correction process on digital data that is stored in a buffer memory and includes multiple code words. The apparatus includes a memory access circuit connected to the buffer memory for reading and writing the code words to the buffer memory in accordance with an address signal. An address generation circuit generates the address signal to designate an address of the code words read from and written to the buffer memory and to provide the address signal to the memory access circuit. A timing generation circuit is connected to the address generation circuit for generating a timing signal to control the address generation circuit so that the address signal is generated to read the code words from the buffer memory. A syndrome generation circuit generates multiple syndromes in parallel in correspondence with the multiple code words by processing the code words of the digital data read from the buffer memory. [0015]
  • A further perspective of the present invention is a method for performing an error correction process on digital data that is stored in a buffer memory and includes multiple code words. The method includes consecutively reading the code words of the digital data from the buffer memory, generating multiple syndromes in parallel by processing the code words of the digital data read from the buffer memory, and performing the error correction process on the digital data using the multiple syndromes. [0016]
  • A further perspective of the present invention is a method for performing an error correcLion process on digital data that is stored in a buffer memory and includes multiple code words. The method includes generating an address signal to consecutively read the code words from the buffer memory, consecutively reading the code words of the digital data from the buffer memory in accordance with the address signal, generating multiple syndromes in parallel by processing the code words read from the buffer memory, and performing the error correction process on the digital data using the multiple syndromes. [0017]
  • Other perspectives and advantages of the present invention will become apparent from the following description, taken in conjunction with the accompanying drawings, illustrating by way of example the principles of the invention.[0018]
  • BRIEF DESCRIPTION OF THE DRAWINGS
  • The invention, together with objects and advantages thereof, may best be understood by reference to the following description of the presently preferred embodiments together with the accompanying drawings in which: [0019]
  • FIGS. [0020] 1(a) to 1(d) are diagrams illustrating the data format of a prior art CD-ROM;
  • FIG. 2 is a diagram illustrating a code word of a prior art CD-ROM; [0021]
  • FIG. 3 is a diagram illustrating the prior art CD-ROM data stored in a DRAM; [0022]
  • FIG. 4 is a diagram illustrating the order for reading the prior art code word; [0023]
  • FIG. 5 is a diagram illustrating the order for reading a code word in the present invention; [0024]
  • FIG. 6 is a schematic block diagram illustrating an error correction apparatus according to one embodiment of the present invention; [0025]
  • FIG. 7 is a schematic circuit diagram of a syndrome generation circuit of the error correction apparatus of FIG. 6; [0026]
  • FIG. 8 is a schematic block diagram of a symbol counter of the error correction apparatus of FIG. 6; [0027]
  • FIGS. [0028] 9(a) to 9(o) are time charts illustrating the timing for reading a code word of the error correction apparatus of FIG. 6;
  • FIG. 10 is a time chart illustrating the timing for reading the prior art code word; and [0029]
  • FIG. 11 is a time chart illustrating the timing for reading a further code word of the error correction apparatus of FIG. 6.[0030]
  • DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
  • In the drawings, like numerals are used for like elements throughout. [0031]
  • A preferred embodiment of an error correction apparatus according to the present invention will now be discussed with reference to the drawings. The error correction apparatus corrects errors using P codes of CD-ROM. [0032]
  • Referring to FIG. 5, in the present invention, four code words are consecutively read from the DRAM when reading P codes. In other words, for the data of the MS bytes and the data of the LS bytes, the data corresponding to data numbers “i+43i”, “(i+1)+43j”, “(i+2)+43j”, “(i+3)+43j”, (i=0, 1, . . . 39: j=0, 1, . . . 25)” are read consecutively. With each of the four code words, when all of the symbols of the 26 bytes of “j=0 to 25” is read, the reading of the next four code words starts. [0033]
  • Each symbol of the four code words, which are adjacent to one another, is stored in the DRAM according to the adjacent address. Therefore, after designating a row address of the DRAM, a page mode technique, which changes the column addresses in accordance with each code word, is employed to read data from the DRAM. This requires only six clock signals to read four symbols. That is, one clock signal is allocated to precharge, one clock signal is allocated to row address designation, and four clock signals are allocated to column address designation. In this manner, four code words are read consecutively to reduce the number of clock signals. This enables high speed access. [0034]
  • As shown in FIG. 5, when reading four code words consecutively, after reading 1040 bytes of data of the LS bytes or the MS bytes, only three code words remain. Accordingly, three code words are read consecutively through the page mode technique for the data from byte 1041 to byte 1118. [0035]
  • When reading each symbol of the four code words, the row address is re-designated in an exceptional manner when the row address changes. In the preferred embodiment, the read address is detected to determine whether or not the column address of the DRAM of FIG. 3 is “255”. When it is determined that the column address is “255”, a new row address is designated. [0036]
  • FIG. 6 is a schematic block diagram of an error correction apparaLus [0037] 100 according to a preferred embodiment of the present invention. The error correction apparatus 100 includes a DRAM 10 for storing the CD-ROM data and a bus arbiter 11 for selecting the process that is to be performed in the DRAM 10. The DRAM 10 has the same configuration as that of the DRAM of FIG. 3.
  • The bus arbiter [0038] 11 is provided with the CD-ROM data transferred from a CD signal processing circuit (not shown) and the addresses of the DRAM 10. The bus arbiter 11 generates a row address strobe (RAS) signal and a column address strobe (column address strobe) signal to write data to the DRAM 10 at the designated address. The RAS signal, the CAS signal, and the CD-ROM data are written to the DRAM 10. When reading data from the DRAM 10, the RAS signal and the CAS signal are used to access the data subject to the reading.
  • The error correction apparatus [0039] 100 includes an error timing generation circuit 20 for reading write data in the manner shown in FIG. 5. When the error correction timing generation circuit 20 designates the data that is to he read from the DRAM 10, a sector head address generation circuit 21 designates the byte number corresponding to the head of the sector to which the data read from the DRAM 10 belongs. The head of the sector is designated using the sum of the bytes of data written to the DRAM 10. In other words, the head “0000” of sector 1 is designated when reading data that belongs to sector 1 of FIG. 3. The bead “2352” is designated when reading data belonging to sector 2. Further, the head “4704” of sector 3 is designated when reading data belonging to sector 3.
  • When the error correction Liming generation circuit [0040] 20 designates the data that is to be read from the DRAM 10, an ECC offset address generation circuit 22 designates the data that is subject to reading in accordance with the order in the sector. More specifically, the data subject to reading is basically designated in the order of 110000, 0001, 0002, 0003, 0043, 0044, 0045, 0046, 0086, . . .”. However, since the first 6 bytes of data from the head of each sector (the data of the LS bytes and the data of the MS bytes) are synchronizing data, the six bytes of data are not read to calculate the syndrome. Thus, the ECC offset address generation circuit 22 designates the data that is subject to reading in the order of “0006, 0007, 0008, 0009, 0049, 0050, 0051, 0052, 0092, . . .
  • Based on the byte number designated by the sector head address generation circuit [0041] 21 and the subject reading data designated by the ECC offset address generation circuit 22, an adder 23 calculates a designated value of the data reading order in the DRAM 10. In this case, the storing of data by the memory cell of the DRAM 10 in units of 2 bytes is considered. That is, with respect to the data of sector 1 of FIG. 3, designated values in the reading order of “0006, 0007, 0008, 0009, 0049, 0050, 0051, 0052, 0092, . . .” are calculated. With respect to the data of sector 2, designated values in the reading order of “1181, 1182, 1183, 1184, 1224, 1225, 2226, 2227, 2267, . . . , which are obtained by sequentially adding 1175 bytes, are calculated.
  • When the designated value calculated by the adder [0042] 23 is provided to the bus arbiter 11, the bus arbiter 11 reads data from the DRAM 10 in accordance with the designated values. The designated values of the adder 23 are also provided to a page boundary detection circuit 24. The page boundary detection circuit 24 generates a page boundary signal when the column address of the DRAM 10 of the reading subject data is “255” (page boundary). The boundary is set when the lower eight bits of a binary designated value of the adder 23 are all “1”. By detecting the page boundary in this manner, the error correction timing generation circuit 20 determines whether the row address of the DRAM 10 changes when four (or three) symbols corresponding to four (or three) code words are read.
  • When determining that the row address does not change, the error correction timing generation circuit [0043] 20 provides the bus arbiter 11 with an instruction for designating the row address once and then designating the column address corresponding to each code word of the reading subject four times (or three times) to read the code word. When determining that the row address changed, the error correction timing generation circuit 20 provides the bus arbiter 11 with an instruction for newly designating the row address after reading the data of the page boundary.
  • A CD-ROM decoding circuit [0044] 30 performs an error correction process on the data read from the DRAM 10. The CD-ROM decoding circuit 30 includes a syndrome generation circuit 31, an error correction circuit 32, an ECC address generation circuit 33, and an error detection circuit 34, The syndrome generation circuit 31 is provided with the data read from the DRAM 10 via the bus arbiter 11. An example for correcting an error in the preferred embodiment will now be discussed. The syndrome generation circuit 31 calculates the two syndromes of S0 and S1 for each code word. The syndrome S0 is calculated by performing the operation of S0=S0 v(i) on all of the symbols of the code words. The initial value represents “0”, v(i) represents the value of the ith symbol of the code word, and represents the adding of a Galois body. The syndrome S1 is calculated by performing the operation of SO=(a*SO) v(i) on the values of all of the symbols of the code words. The initial value is represented by “0”, a represents a primitive element, and * represents the multiplying of a Galois body.
  • The syndrome generation circuit [0045] 31 generates the syndromes S0, S1 of each of Lhe four code words in parallel. Referring to FIG. 7, operational circuits S00, S01, S02, S03 calculate the syndrome S0, and operational circuits S10, S11, S12, S13 calculate the syndrome S1. In other words, when a register (represented as reg in FIG. 7) of an operational circuit corresponding to a code word is activated, the operational circuit performs a calculation with the symbol v (i) read froili the DRAM 10. Each operational circuit calculates the syndrome by performing a calculation using the values v (i) (i=0, 1, . . . 25) of the 26 symbols. The syndrome SO is transferred to the error detection circuit 34 via a multiplexer M1, and the syndrome S1 is transferred to the error detection circuit 34 via a multiplexer M2. The error detection circuit 34 determines whether all error code includes an error based on the syndromes S0, S1. If there is an error, the value of the error and the location of the error are calculated, and the error correction circuit 32 and the ECC address generation circuit 33 are provided with the calculated results.
  • The ECC address generation circuit [0046] 33 generates an address designating the location of an error and provides the bus arbiter 11 with the address. The bus arbiter 11 reads data from the erroneous location of the DRAM 10 and provides the read data to the error correction circuit 32. The error correction circuit 32 generates correct data from the erroneous data and the value of the error provided from the error detection circuit 34. The erroneous data of the DRAM 10 is rewritten to the correct data.
  • After the error correction, an EDC decoder block (not shown) checks whether error correction was properly performed using the EDC data of FIG. 1([0047] a) After such series of processes are performed, when a icost computer (not shown) instructs the bus arbiter 11 to transfer data, the bus arbiter 11 reads data from the DRAM 10 and transfers the read data to the host computer.
  • The error correction timing generation circuit [0048] 20 controls the timing for reading code words from the DRAM 10 and the timing for calculating the syndrome with the syndrome generation circuit. Referring to FIG. 8, the error correction timing generation circuit 20 includes a symbol counter 150 for allocating the data selectively read from the DRAM 10 by each operational circuit.
  • The symbol counter [0049] 150, which includes a register 40, a count control circuit 42, and a decoder 41, increments a count value in accordance with, for example, the signal of a CAS or the like (the signal synchronized with the timing for reading data from the DRAM 10). The lower two digits “00”, “01”, “10”, “11” of the count value of the symbol counter 150 respectively correspond to four code words. The data read from the DRAM 10 is selectively distributed to a plurality of operational circuits based on the count value.
  • The register [0050] 40 temporarily holds the count value generated by the count control circuit 42 and provides the count value to the decoder 41 and the count control circuit 42 in synchronism with the timing signal (e.g. CAS signal). Based on the lower two digits of the count value, the decoder 41 selectively provides an enable signal to the registers of the operational circuits S00, S10, the registers of the operation circuits S01, S11, the registers of the operational circuits S02, S12, and the registers of the operational circuits S03, S13. The syndromes of the four code words are calculated by the operational circuits in response to the enable signal.
  • The count control circuit [0051] 42 receives a count value from the reqister 40, adds “1”or “2” to the count value, and provides the added count value to the register 40. When the count value is “0000” to “1039”, “1” is added to the count value (i.e., the count value is incremented) When the count value reaches “1040”, “1” or “2” is added to the count value based on the lower two digits of the count value. When the lower two digits of the count value is “0” or “01”, “1” is added to the count value. When the lower two digits of the count value is “10”, “2” is added to the count value to set the lower two digits of the count value to “00”. By controlling the count value, the lower two digits of the count value is prevented from being “11” when the count value is “1040” or greater.
  • When the symbols of 1040 bytes are read, only three code words remain as shown in FIG. 5. Thus, in synchronism with the timing for reading data from the DRAM [0052] 10 and the timing of the three code words, the registers of the operational circuits S00, S10, the registers of the operational circuits S01, S11, and the registers of the operational circuits S02, S12 are selectively activated to calculate the syndromes of the three code words in parallel.
  • When [0053] 26 symbols are distributed to each operational circuit and the syndrome of each code word is calculated, the error correction timing generation circuit 20 generates a clear signal to initialize the operational circuit. By providing the operation circuit with the clear signal, the operational circuit generates a new syndrome based on the symbol.
  • The error correction timing generation circuit [0054] 20 includes a counter for providing a reset signal to the symbol counter 150 of FIG. 8 when determining that a sector of data has been read from the DRAM 10. In this manner, the counter value of the symbol counter 150 is initialized when it is determined that a secLor of data has been read from the DRAM 10. Accordingly, the enable signal is generated without causing the configuration of the symbol counter 150 to be complicated.
  • The timing for accessing the DRAM [0055] 10 and the timing for calculating the syndrome will now be discussed with reference to FIGS. 9(a) to 9(o).
  • When the error correction timing generation circuit [0056] 20 provides the syndrome generation circuit 31 with a clear signal (FIG. 9(f)) in synchronism with a predetermined clock signal (FIG. 9(a)), the operational circuits S00, S10, the operational circuits S01, S11, the operational circuits S02, S12, and the operational circuits S03, S13 are initialized. This enables each operation circuit to perform a new syndrome calculation.
  • Then, to designabe the address of the data read from the DRAM [0057] 10, the RAS signal (FIG. 9(c)) and the CAS signal (FIG. 9(d)) are generated by the bus arbiter 11. The row address of the DRAM 10 is designated when the RAS signal goes low, and the coliimn address is designated when the CAS signal goes low. The designated row and column addresses are held until the RAS signal and the CAS signal go low the next time
  • The RAS signal is maintained at a constant level (low level) (i.e., the row address is maintained) until the CAS signal designates four column addresses Accordingly, the data of four code words (d[0058] 0, d1, d2, d3) is read (FIG. 9(e)) in synchronism with the trailing edge of the CAS signal. The error correction timing generation circuit 20 generates an enable signal (FIG. 9(g)) and calculates the syndrome with the operational circuit of the syndrome generation circuit 31 using the read data. That is, the enable signal processes the data d0 with the operational circuits S00, S10 (FIGS. 9(h) and 9(i)), the data d1 with the operational circuits S01 and S11 (FIGS. 9(j) and 9(k)), the data d2 with the operational circuits S02, S12 (FIGS. 9(j) and 9(m)), and the data d3 with the operational circuits S03, S13 (FIGS. 9(n), 9(o)).
  • After reading the data d[0059] 0, d1, d2, d3, the RAS signal and the CRS signal go high, and the DRAM 10 performs a precharge operation (as indicated by xx in FIG. 9(b)). After the precharge operation, the RAS signal and the CAS signal go low again, and a predetermined address of the DRAM 10 is accessed to read data.
  • In this manner, four pieces of data are read with 6 clock signals by designating the column address to which the four code words belong after designating the row address. In comparison, in the prior art method of FIG. 4 in which data is read for one code word at a time, the DRAM [0060] 10 is accessed as shown in FIG. 10. More specifically, each piece of data that is subject to reading requires three (3) clock signals, those for designating the row address and the column address of the DRAM 10 with the RAS signal and the CAS signal and that for performing the precharge operaLion (indicated as xx). Accordingly, when reading four pieces of data, 12 clock signals are necessary. Thus, the time for reading four pieces of data is twice as long in comparison with the preferred embodiment.
  • As shown in the example of FIG. 11, if two column addresses are designaled by the CAS signal in the preferred embodiment, the page boundary detection circuit [0061] 24 generates a page boundary signal SB when the column address indicates a page boundary of the DRAM 10. Tn response to the page boundary signal SB, the RAS signal and the CAS signal go high and the precharge operation (indicated by xx) is performed. Then, after designating a row address again with the RAS signal, the CAS signal designates a column address to read the two pieces of data d2, d3.
  • The error correction apparatus of the preferred embodiment has the advantages discussed below. [0062]
  • (1) A plurality of P codes are consecutively read from the DRAM [0063] 10. Thus, after designating the row address, the page mode technique, which consecutively designates column addresses, may be applied. This shortens the time the DRAM 10 is accessed.
  • (2) The page boundary detection circuit [0064] 24 detects whether the data subject to reading is at the boundary of a page. This guarantees that the processing based on the page mode technique is performed.
  • (3) The syndrome generation circuit [0065] 31 processes four code words in parallel. Thus, syndrome calculation is performed within a short period of time.
  • (4) After the 1040 MS bytes and 1040 LS bytes are read, the symbol counter provides the enable signal only to the operational circuits S[0066] 00, S10, the operational circuits S01, S11, and the operational circuits S02, S12. This guarantees the reading of the code word even it the total number of code words is not a multiple of the number of code words that are read consecutively.
  • (5) The multiplexers M[0067] 1, M2 distribute the error detection circuit 34 with syndromes, which are generated by processing four code words in parallel. This prevents the circuit area of the error detection circuit 34 from increasing.
  • It should be apparent to those skilled in the art that the present invention may be embodied in many other specific forms without departing from the spirit or scope of the invention. Particularly, it should be understood that the present invention may be embodied in the following forms. [0068]
  • (a) The page size of the DRAM [0069] 10 is not limited to 256 bytes. When changing the page size, the page boundary detected by the page boundary detection circuit 24 is also changed.
  • (b) Instead of the symbol counter, a read only memory (ROM) may provide the enable signal. [0070]
  • (c) The application of the present invention is not limited to consecutive reading of the P codes. The present invention may also be applied to consecutive reading of the Q codes. In such a case, when Lhe Q codes are first read, it is preferred that the data of a single word code be first read and that the data of multiple code words be read afterward. More specifically, referring to FIG. 2, it is preferred that “0000” be first read, “0043” and “0044” be read next, and “0086”, “0087”, and “0088” be read afterward. In the subsequent processing, the data of four code words (e.g., “0129”, “1130”, “0131”, and “0132”) are read consecutively. In such case, the process for providing the syndrome generation circuit [0071] 31 with the enable signal and the clear signal is changed as required.
  • (d) The application of the invention is not restricted to the error correction process of a CD-ROM. The present invention may also be applied to an error correction process of a CD or a digital versatile disc (DVD). [0072]
  • (e) The number of code words read consecutively and the number of syndromes calculated in parallel are not restricted as described in the preferred embodiment. Further, the symbol of code words read consecutively does not have to be stored adjacent to a random access memory. [0073]
  • (f) When consecutively reading plural code words, a high speed access technique, such as the extended data output (EDO) technique, may be employed in lieu of the page mode technique. Further, a synchronous DRAM (SDRAM) may be used in lieu of the DRAM. [0074]
  • (g) The multiplexers M[0075] 1 and M2 may be eliminated.
  • The present examples and embodiments are to be considered as illusLraLive and riot restrictive, and the invention is not to be limited to the details given herein, but may be modified within the scope and equivalence of the appended claims. [0076]

Claims (16)

    What is claimed is:
  1. 1. An error correction apparatus for performing an error correction process on digital data that is stored in a buffer memory and includes multiple code words, the apparatus comprising:
    a memory access circuit for controlling reading and writing of the code words to the buffer memory; and
    a plurality of operational circuits for pertorming a syndrome calculalioii with each of the multiple code words read from the buffer memory, wherein the memory access circuit consecutively reads the multiple code words from the buffer memory and distributes the read code words to the plurality operational circuits.
  2. 2. The apparatus according to claim 1, wherein each of the code words is formed by multiple symbols, and the memory access circuit reads the symbols of the code words.
  3. 3. The apparatus according to claim 2, wherein the memory access circuit includes:
    a register for holding a count value indicating the number of times the symbols of the code words are read from the buffer memory;
    a decoder for distributing the symbols read from Llie buffer memory to the plurality of the operational circuits in accordance with the count value; and
    a count control circuit for altering the count value in accordance with a timing for reading the symbols.
  4. 4. The apparatus according to claim 3, wherein the count control circuit changes an adding condition of the count value when the count value exceeds a predetermined value.
  5. 5. The apparatus according to claim 1, wherein the memory access circuit holds a row address of the buffer memory and changes a column address for each code word to consecutively read the multiple code words from the buffer memory.
  6. 6. An error correction apparatus for performing an error correction process on digital data that is stored in a buffer memory and includes multiple code words, the apparatus comprising:
    a memory access circuit connected to the buffer memory for reading and writing the code words to the buffer memory in accordance with an address signal;
    an address generation circuit tor generating the address signal to designate an address of the code words read from and written to the buffer memory and for providing the address signal to the memory access circuit;
    a timing generation circuit connected to the address generation circuit for generating a timing signal to control the address generation circuit so that the address signal is generated to read the code words from the buffer memory; and
    a syndrome generation circuit for generating multiple syndromes in parallel in correspondence with the multiple code words by processing the code words of the digital data read from the buffer memory.
  7. 7. The apparatus according to claim 6, wherein the address signal includes a row address signal and a column address signal, and the control circuit controls the address generation circuit so that a plurality of the column address signals corresponding to the multiple code words are consecutively generated after the row address signal is generated.
  8. 8. The apparatus according to claim 6, further comprising:
    a detection circuit for detecting a page boundary signal by detecting a page boundary of the digital data based on the address signal generated by the address generation circuit.
  9. 9. The apparatus according to claim 8, wherein the timing generation circuit generates the timing signal to control the address generation circuit so that the address signal changes in accordance with the page boundary signal.
  10. 10. The apparatus according to claim 6, wherein the syndrome generation circuit includes a plurality of operational circuits for generating the syndromes in parallel.
  11. 11. A method for performing an error correction process on digital data that is stored in a buffer memory and includes multiple code words, the method comprising the steps of:
    consecutively reading the code words of the digital data from the buffer memory;
    generating multiple syndromes in parallel by processing the code words of the digital data read from the buffer memory; and
    performing the error correction process on the digital data using the multiple syndromes.
  12. 12. The method according to claim 11, wherein the step for generating multiple syndromes includes generating the syndromes using a plurality of operational circuits.
  13. 13. A method for performing an error correction process on digital data that is stored in a buffer memory and includes multiple code words, the method comprising the steps of:
    consecutively generating an address signal to read the code words of the digital data from the buffer memory;
    consecutively reading the code words of the digital data from the buffer memory in accordance with the address signal;
    generating multiple syndromes in parallel by processing the code words of the digital data read from the buffer memory; and
    performing the error correction process on the digital data using the multiple syndromes.
  14. 14. The method according to claim 13, wherein the address signal includes a row address signal and a column address signal, and the step for generating an address signal includes generating a row address signal and then consecutively generating a plurality of the column address signals corresponding to the multiple code words.
  15. 15. The method according to claim 13, further comprising the step of;
    generating a page boundary signal by detecting a page boundary of the digital data based on the address signal.
  16. 16. The method according to claim 15, further comprising the step of:
    changing the address signal in accordance with the page boundary signal.
US10105010 2001-03-22 2002-03-22 Error correction apparatus for performing consecutive reading of multiple code words Expired - Fee Related US7143331B2 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
JP2001082298A JP3954803B2 (en) 2001-03-22 2001-03-22 Error correction device
JP2001-082298 2001-03-22

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
US11552373 US20070050663A1 (en) 2001-03-22 2006-10-24 Error correction apparatus for performing consecutive reading of multiple code words

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US11552373 Continuation US20070050663A1 (en) 2001-03-22 2006-10-24 Error correction apparatus for performing consecutive reading of multiple code words

Publications (2)

Publication Number Publication Date
US20020144206A1 true true US20020144206A1 (en) 2002-10-03
US7143331B2 US7143331B2 (en) 2006-11-28

Family

ID=18938258

Family Applications (2)

Application Number Title Priority Date Filing Date
US10105010 Expired - Fee Related US7143331B2 (en) 2001-03-22 2002-03-22 Error correction apparatus for performing consecutive reading of multiple code words
US11552373 Abandoned US20070050663A1 (en) 2001-03-22 2006-10-24 Error correction apparatus for performing consecutive reading of multiple code words

Family Applications After (1)

Application Number Title Priority Date Filing Date
US11552373 Abandoned US20070050663A1 (en) 2001-03-22 2006-10-24 Error correction apparatus for performing consecutive reading of multiple code words

Country Status (3)

Country Link
US (2) US7143331B2 (en)
JP (1) JP3954803B2 (en)
KR (1) KR100509137B1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070050663A1 (en) * 2001-03-22 2007-03-01 Sanyo Electric Co., Ltd. Error correction apparatus for performing consecutive reading of multiple code words
US20080145064A1 (en) * 2006-12-13 2008-06-19 Masaki Ohira Optical line terminal and optical network terminal

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7698625B2 (en) * 2003-08-26 2010-04-13 Adaptec, Inc. System for improving parity generation and rebuild performance
US7228490B2 (en) * 2004-02-19 2007-06-05 Quantum Corporation Error correction decoder using cells with partial syndrome generation
US7343546B2 (en) * 2004-12-23 2008-03-11 Intel Corporation Method and system for syndrome generation and data recovery
JP2006190346A (en) * 2004-12-28 2006-07-20 Toshiba Corp Error correction processing device and error correction processing method
JP2006309820A (en) 2005-04-26 2006-11-09 Sanyo Electric Co Ltd Error correction device

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5612922A (en) * 1995-07-05 1997-03-18 Micron Technology, Inc. Page mode editable real time read transfer
US5721860A (en) * 1994-05-24 1998-02-24 Intel Corporation Memory controller for independently supporting synchronous and asynchronous DRAM memories
US6044484A (en) * 1997-04-10 2000-03-28 Samsung Electronics Co., Ltd. Method and circuit for error checking and correction in a decoding device of compact disc-read only memory drive
US6158038A (en) * 1996-11-15 2000-12-05 Fujitsu Limited Method and apparatus for correcting data errors
US6185629B1 (en) * 1994-03-08 2001-02-06 Texas Instruments Incorporated Data transfer controller employing differing memory interface protocols dependent upon external input at predetermined time
US6243845B1 (en) * 1997-06-19 2001-06-05 Sanyo Electric Co., Ltd. Code error correcting and detecting apparatus
US6332206B1 (en) * 1998-02-25 2001-12-18 Matsushita Electrical Industrial Co., Ltd. High-speed error correcting apparatus with efficient data transfer
US6591349B1 (en) * 2000-08-31 2003-07-08 Hewlett-Packard Development Company, L.P. Mechanism to reorder memory read and write transactions for reduced latency and increased bandwidth
US6651208B1 (en) * 2000-04-04 2003-11-18 Mosel Vitelic Corporation Method and system for multiple column syndrome generation

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6687860B1 (en) * 1998-01-07 2004-02-03 Matsushita Electric Industrial Co., Ltd. Data transfer device and data transfer method
JP3954803B2 (en) * 2001-03-22 2007-08-08 三洋電機株式会社 Error correction device

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6185629B1 (en) * 1994-03-08 2001-02-06 Texas Instruments Incorporated Data transfer controller employing differing memory interface protocols dependent upon external input at predetermined time
US5721860A (en) * 1994-05-24 1998-02-24 Intel Corporation Memory controller for independently supporting synchronous and asynchronous DRAM memories
US5612922A (en) * 1995-07-05 1997-03-18 Micron Technology, Inc. Page mode editable real time read transfer
US6158038A (en) * 1996-11-15 2000-12-05 Fujitsu Limited Method and apparatus for correcting data errors
US6044484A (en) * 1997-04-10 2000-03-28 Samsung Electronics Co., Ltd. Method and circuit for error checking and correction in a decoding device of compact disc-read only memory drive
US6243845B1 (en) * 1997-06-19 2001-06-05 Sanyo Electric Co., Ltd. Code error correcting and detecting apparatus
US6332206B1 (en) * 1998-02-25 2001-12-18 Matsushita Electrical Industrial Co., Ltd. High-speed error correcting apparatus with efficient data transfer
US6651208B1 (en) * 2000-04-04 2003-11-18 Mosel Vitelic Corporation Method and system for multiple column syndrome generation
US6591349B1 (en) * 2000-08-31 2003-07-08 Hewlett-Packard Development Company, L.P. Mechanism to reorder memory read and write transactions for reduced latency and increased bandwidth

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070050663A1 (en) * 2001-03-22 2007-03-01 Sanyo Electric Co., Ltd. Error correction apparatus for performing consecutive reading of multiple code words
US20080145064A1 (en) * 2006-12-13 2008-06-19 Masaki Ohira Optical line terminal and optical network terminal
US7978972B2 (en) * 2006-12-13 2011-07-12 Hitachi, Ltd. Optical line terminal and optical network terminal

Also Published As

Publication number Publication date Type
KR20020075269A (en) 2002-10-04 application
KR100509137B1 (en) 2005-08-23 grant
US20070050663A1 (en) 2007-03-01 application
US7143331B2 (en) 2006-11-28 grant
JP2002280909A (en) 2002-09-27 application
JP3954803B2 (en) 2007-08-08 grant

Similar Documents

Publication Publication Date Title
US7162678B2 (en) Extended error correction codes
US6639865B2 (en) Memory device, method of accessing the memory device, and reed-solomon decoder including the memory device
US6223322B1 (en) Method and apparatus for enhancing data rate in processing ECC product-coded data arrays in DVD storage subsystems and the like
US6661591B1 (en) Disk drive employing sector-reconstruction-interleave sectors each storing redundancy data generated in response to an interleave of data sectors
US5459742A (en) Solid state disk memory using storage devices with defects
US4667326A (en) Method and apparatus for error detection and correction in systems comprising floppy and/or hard disk drives
US4675869A (en) Fast decoder and encoder for Reed-Solomon codes and recording/playback apparatus having such an encoder/decoder
US5878059A (en) Method and apparatus for pipelining an error detection algorithm on an n-bit word stored in memory
US5623506A (en) Method and structure for providing error correction code within a system having SIMMs
US4763332A (en) Shared circuitry for the encoding and syndrome generation functions of a Reed-Solomon code
US6353910B1 (en) Method and apparatus for implementing error correction coding (ECC) in a dynamic random access memory utilizing vertical ECC storage
US5428630A (en) System and method for verifying the integrity of data written to a memory
US4450561A (en) Method and device for generating check bits protecting a data word
US5157669A (en) Comparison of an estimated CRC syndrome to a generated CRC syndrome in an ECC/CRC system to detect uncorrectable errors
US5996105A (en) ECC system employing a data buffer for storing codeword data and a syndrome buffer for storing error syndromes
US6973613B2 (en) Error detection/correction code which detects and corrects component failure and which provides single bit error correction subsequent to component failure
US6691205B2 (en) Method for using RAM buffers with simultaneous accesses in flash based storage systems
US6976194B2 (en) Memory/Transmission medium failure handling controller and method
US6216247B1 (en) 32-bit mode for a 64-bit ECC capable memory subsystem
US5027357A (en) ECC/CRC error detection and correction system
US20050132259A1 (en) Error correction method and system
US5247523A (en) Code error correction apparatus
US6009547A (en) ECC in memory arrays having subsequent insertion of content
US6470473B1 (en) Data decoding processing system and method
US5666371A (en) Method and apparatus for detecting errors in a system that employs multi-bit wide memory elements

Legal Events

Date Code Title Description
AS Assignment

Owner name: SANYO ELECTRIC CO., LTD., JAPAN

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:TOMISAWA, SHIN-ICHIRO;REEL/FRAME:012893/0627

Effective date: 20020319

REMI Maintenance fee reminder mailed
LAPS Lapse for failure to pay maintenance fees
FP Expired due to failure to pay maintenance fee

Effective date: 20101128